Smyths Toys deploys Codie, a Gemini multi-agent customer service system on SAP, handling about 60% of peak customer service workload
Smyths Toys, an Irish toy retailer, launched Codie in January 2025 to answer 'Where's my order?' enquiries and expanded it across email, web and social media, using Gemini, Google's Agent Development Kit and the Google Cloud ABAP SDK with live SAP S/4HANA data. The source reports about 60% of peak customer service workload handled, 90%+ customer satisfaction, more than 300,000 emails resolved over six months and over $1.3 million saved in temporary staffing costs over 12 months.
Overview
Smyths Toys, an Irish toy retailer, launched Codie in January 2025 to answer 'Where's my order?' enquiries and expanded it across email, web and social media, using Gemini, Google's Agent Development Kit and the Google Cloud ABAP SDK with live SAP S/4HANA data. The source reports about 60% of peak customer service workload handled, 90%+ customer satisfaction, more than 300,000 emails resolved over six months and over $1.3 million saved in temporary staffing costs over 12 months.
This entry has 11 published fields tied to exact passages in an immutable source capture.
Inspect the highlighted sourceThe challenge
Every holiday season Smyths' customer service was inundated with thousands of enquiries, mostly "Where's my order?"; each required a team member to manually retrieve information from multiple systems, and during peak periods customers waited as long as 48 hours for an email response.
The solution
Smyths built Codie, which launched as a proof of concept in January 2025 to answer "Where's my order?" enquiries. A supervisor agent plans how each enquiry is resolved and delegates to specialised sub-agents that use Gemini function calling to retrieve live data from SAP S/4HANA, carrier APIs and payment platforms. SAP processes run through the Google Cloud ABAP SDK, while AI workloads run in Python on Kubernetes using Google's Agent Development Kit.
Reported business value
The source reports that at peak Codie handled approximately 60% of the customer service workload with customer satisfaction above 90%, resolved more than 300,000 customer emails over six months, and saved more than $1.3 million in temporary staffing costs over 12 months.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other retail entries in the register.
John Keells Holdings scales retail advanced analytics with Azure Databricks
John Keells Holdings' OCTAVE Center of Excellence used Azure Databricks, Delta Lake and MLflow to unify siloed data across 70 companies and seven industry verticals, running over 40 advanced analytics use cases including a store-clustering supply chain optimization pilot for its Keells supermarket chain that improved dry-category revenue and cut the time for annual category review.
Omoda's generative AI stylist lifts conversion rate 2.5x for online fashion shoppers
Dutch fashion retailer Omoda built AI stylist, a virtual shopping assistant powered by Google Cloud's Gemini Pro 1.5 model with Vertex AI Multimodal Embeddings and Vector Search. AI stylist holds natural-language conversations with customers, understands their needs, and builds complete outfits combining clothing, shoes, and accessories rather than just matching similar products, with an Intelligent Brand Matching feature that factors in brand affinities. Customers who use AI stylist convert at a rate 2.5 times higher than average, and the tool also improves inventory management through better trend anticipation.
Furniture.com Transforms Online Search with Databricks
Furniture.com unifies over 60 retail partners and 1.5 million SKUs on the Databricks Data + AI Platform, using Delta Lake, MLflow and Unity Catalog to run its ML lifecycle. Its Find It AI product-discovery tool uses generative AI to create a synthetic image representing shopper intent, then matches it against the product catalog for image-based search. A Collections model uses LLMs to automatically group related products, finding more than 16,000 collections across 50+ partners with no human intervention. Users who interact with Find It AI show a click-through rate 8x higher than baseline and a return rate 3.2x higher than baseline.
How Walmart Achieved Enterprise Transformation Through Digital Experience Innovation
Working with implementation partner Nativa on LivePerson's technology, Walmart customized and integrated conversational technology into its WhatsApp channel, including a FAQ-Transactional Bot to streamline searches, inquiries and purchases. The solution delivered a 60% increase in productivity and a 10 percentage point improvement in CSAT/NPS, along with operational savings through automated customer service processes.
Was this helpful?
Your feedback helps us improve our use case database

